Collette Louise English BMBS, FRCR, Presenter: Nothing to Disclose

Mark Teh MBBS, Abstract Co-Author: Nothing to Disclose

Rebecca E. Hall MBChB, Abstract Co-Author: Nothing to Disclose

Ana-Maria Bilawich MD, Abstract Co-Author: Nothing to Disclose

Stephen Lam, Abstract Co-Author: Nothing to Disclose

John R. Mayo MD, Abstract Co-Author: Speaker, Siemens AG

We studied the utility of a nodule follow up algorithm based on malignancy risk assessed using a published Nodule risk calculator applied to lung cancer screening low dose CT (LDCT) scans. Using this approach, we hypothesized more appropriate CT follow up could be performed, reducing cost and radiation dose. 

We performed a retrospective study of a cross sectional sample of 200 consecutive LDCT scans (February  to May 2003) obtained in a lung cancer screening program. Entry criteria for screening included; 30 pack year current or former smoker, 50-74, no previous lung cancer. Baseline CT was assessed and dominant nodule malignant risk assessed using the nodule calculator based on; nodule diameter, nodule density (solid, part solid, ground glass), age, spiculation, lobar location, gender, family history of lung cancer, emphysema and total nodule count. Up to 12 year follow up (FU) was evaluated for; number of follow up LDCT, diagnosis of lung cancer, total radiation dose and fatality from lung cancer. Comparison was made to proposed follow up based on nodule calculator; <1.5% 2 year FU (minimal risk), 1.5-6% 1 year FU (low risk), >6% FU as clinically indicated (high risk).    

Percentage of subjects stratified by dominant nodule risk: 82%with minimal risk; 12% low risk; 6% high risk . Lung cancer developed in 5 participants (3%) in the minimal and low risk group, only one of which arose from the dominant nodule on baseline LDCT. The total number of follow up LDCTs performed in the minimal and low risk group using conventional FU was 932. Using the risk calculator FU protocol 608 (65%) fewer LDCT scans (average 3 per subject) would have been performed with potential saving of ~$150,000. Using this FU protocol all cancers in the minimal and low risk group would have been detected at T1A size (<20 mm diameter).  

The Nodule Calculator accurately identified minimal and low risk subjects in a lung cancer screening cohort with up to 12 year FU. A modified FU protocol based on the nodule calculator risk would have detected all cancers at T1A stage, saving 65% of  LDCT scans, reducing program costs and subject radiation dose.   

Utilization of the nodule risk calculator can greatly reduce the number of follow up CTs in a lung cancer screeing program. 

English, C, Teh, M, Hall, R, Bilawich, A, Lam, S, Mayo, J, Nodule Risk Calculator: A New Tool to Reduce Low Dose CT Scans in Lung Cancer Screening.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14006668.html